New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Are we sure he did chargeback first? In the ticket sequence it's mentioned after the refund is denied, and I can't find anywhere in thread it's explicitly stated (by op or provider) that chargeback was first.
That’s not really how the 95th percentile rule works. I had ChatGPT calculate the max overage for a month, it comes out to about 45,000 NOK (~$4,400).
The above assumes a 5-minute sampling interval, but we can only guess how Gigahost actually calculates their bills, since it’s not stated anywhere.
Yeah, no.
If it was a $10-20 charge, ok, sure. But with $800 being charged based on a ToS change the customer never knew about nor had enough time to review it, you best believe any sane person would immediately issue a chargeback and let the provider figure shit out themselves.
I assume, most of the tickets it's cut out, we don't even know what was in the ticket. I don't see user mentioning lack of TOS anywhere.
But we can him mentioning chargebacks:
00:35 chargeback mentioned. [2nd message?]
00:41 gigahost says "team will manually check and contact you" [3rd message?]
last message in topic confirmation of chargeback by gigahost [6th message?]
So I don't think he gave gigahost any time to manually check that (-:
I disagree. If provider is communicative it's would be much faster to get a refund (1-2 days of talking) than waiting for PayPal to do the thing [if provider not responding like 14 days?]. Not even going with bank one as this one is even longer.
Totally different thing when it's a mass issue scam and you want to 'alarm' PayPal that fuck me this is some illegal shit or when provider is not responding and ignoring/deadpooled.
I don't wanna get too silly with this cuz we'll find out eventually, but of the ticket snips we see gigahost is clearly saying no refund 1st, then op mentions paypal & bank. In fairness that doesn't mean he hadn't already hit up paypal and just not said it in the ticket, but..
Anyway we'll find out more I'm sure.
You used more than you paid for so you got charged extra simple as that really its an expensive mistake to make
Sure, there should’ve been a warning or maybe some kind of cap or throttle once you hit your limit. But if you get a $800 overage charge then this wasnt just a little bit over it was way way over what you agreed to/committed to. That bandwidth still got used, and @gigahost had to pay for it
It sucks but the mistake is on you @anon505.
Should I worry about my 7$ Gigahost deal (non Ryzen) creating a similar overcharge?
Personally I'd prefer you not worry about it then open a thread when it happens.
Thanks for your service.
We need precise information from @gigahost with respect to how bandwidth is calculated. Using their info we may design firewalls and avoid such invoices.
It's on both sides. There's no where in the ToS that mentioned overages, prices, any information of the current balance/what would be charged. It was clear there's fault on both sides and it's clearer that Gigahost knows this by updating their ToS while this thread is going on.
Yes i agree, but it comes down to common sense when you buy a 1g 95th service and use way over that its not going to be free is it? I dont understand how you can not notice such huge overuse on your own service and then be surprised when you get hit with a bill for it.
There is fault both sides but its just huge stupidity from OP to not notice this or expect no charge
While this is only partially related to this drama, IMHO charging overages (especially for cheap offers) is problematic. For example, my Amsterdam server from @gigahost for $7/yr is connected to a 100G uplink, so it's conceivable that a misconfiguration or a security vulnerability (which could cause it to get hacked) causes 20G traffic 24/7 for a month, and therefore, overage charges around $25000 (of course this isn't likely, but theoretically possible). High overage charges could (as Gigahost, IIRC, calculates in+out traffic) even be caused without the user being at fault, for example due to a DDoS attack that wasn't filtered. Possibly even shutting down the server (that may be possible with the API as soon as one gets near the limit) wouldn't help for incoming traffic (depending on their implementation), so at the moment, I don't see any way to reliably protect oneself from potentially insanely high overage charges, which does make using Gigahost (except for unlimited plans) risky, especially as they, according to their TOS, use some kind of collection services.
Of course, overage charges can, depending on the usecase, be better than suspension or throttling, but I don't think that adding a checkbox to suspend/throttle instead of billing the overages would be that hard, and something like this could have prevented this.
To avoid saying entirely I will say it's mostly on the provider side. Fact is, if the typical cloud/server/vps customer would be expected to know what 95th% is or to even read the TOS. The whole cloud/server/vps industry would collapse within hours.
For EU/EEA there are so many laws/regulations/directives protecting customers that the typical provider does not even have an idea about, but still they think they are capable handling the TOS and everything themselves ending up with things that are legally not even applicable. Even worse when they actually act on such assumptions they make.
How can one even notice it? It was unmetered, without any price set in terms. The client simply relaxed because there was no fee.
Again: why expect a charge if there is nothing stated? Maybe he was expecting a throttle, or a notification to set some bandwidth firewall afterwards.
Ballpark figures since I'm lazy:
800Mbps * 30 days = 259TB
800$/.50$ per Mbps over limit = 1600Mbps
1600Mbps over + 800Mbps included = 2400Mbps
2400 Mbps * 30 days = ~ 777TB
777 - 259 = ~518TB over
One could argue the same for gigahost. They let it get way out of hand. If they're allowing a 7$ server to push 3/4 of a petabyte and expecting someone will just casually pay an 800$ bill that's just foolish on their part. If OP is responsible for monitoring their systems, so is giga responsible for monitoring theirs.
I would personally notice and im sure alot of people would notice aswell, even if you mildy monitor your servers, I have no idea what he was using it for but the traffic increase would of been an insane jump from this a previous period to go OVER by 500TB~ im sure you would notice this in some way shape of form even if you were relaxed about it being unmetered
On one hand, it is a mutual process. GH should have put systems in place (throttling, suspending, notifications) to prevent such a big bill and customers should monitor traffic usage etc. On the other hand, it is in the spirit of good customer service to prevent customer incurring large bills, especially if a customer running $7-10-12/y service with GH.
On my other servers, I have gone above my traffic limit a few times. In one case, server was suspended until the traffic was reset, which was fine. In other cases, the port speed has been throttled, which was fine as well.
Plus, I guess it is always possible to open a ticket and discuss your needs with the provider.
Exactly. Customer was relaxed because there was no rule for using over 500TB. Customer did not expect a huge invoice and a sudden change of terms (to justify that invoice) without any prior notification. There was no background for such invoice, therefore customer was relaxed thinking he can use the whole network since there was no consequences stated.
Now we can debate over consequences and unmetered expectations as much as we like, but the mater of fact is there was no background for that invoice, since there were no rules set in terms.
Thats crazy ignorant to think you can use 500TB+ for free and it will go unnoticed
In my opinion Flux panel of Gigahost needs some adjustments, to limit bandwidth. Possible options when bandwidth limit is reached could be some choice like this:
I don't know how hard this would be to implement.
That's the point: not go unnoticed. This would force a notification or announcement from provider. This notification would force some limitations like throttling, or even suspension. But instead provider also relaxed and issued an invoice of $800 without any terms for it. We're talking of a well established provider here, who should have seen this coming at some point. This was a ticking time bomb that was bound to explode at some point with so many abusers online (I'm surprised it did not happen earlier).
We need clear limitations and clear information on how bandwidth is calculated and valued, so that we know exactly what we pay and why.
I've read the whole thread, I still don't understand how the 95th percentile works. I personally think it's kind of scummy of a provider to let a user use so much with minimal notifications and being unclear about it and then bill them for like $1000 on a $7/y service.
Also, can somebody tell me what's the appeal of this percentile calculation over a clear bandwidth quota like 25TB/month?
If they would implement this then they would lose their extra income from the overused bandwidth. So probably they don't want to limit it.
To explain in simple terms I'll use an analogy. You have water in your house and you have a pipe; so think of the water pipe as the network pipe.
You have a large barrel that you have to fill with water. You fill that barrel but it takes you 5 hours to do it. You're charged per liter spent.
If you live in some top tourist area when there's a lot of turists you will need 20 hours to fill your barrel. That's shared infrastructure.
Now imagine you need to fill your barrel not in 5 hours but in 5 seconds. You need a bigger pipe and if that pipe is shared you need some guarantees. Do you think you will get the same price per liter?
Hello everyone,
We take this matter seriously. As one of the more established and reputable providers on this board we should have acted responsibly.
Our systems already send notifications for bandwidth overuse, but we realise there is always room to improve. No client should have to worry about unexpectedly large charges if a misconfiguration or similar issue occurs.
That is why we quickly developed a feature in Flux to allow for actions to run if you were to exceed current limits. You can either stop the server, suspend it or soon limit it (VPS only). This feature set is now available for all clients from today.
We welcome the OP to retract his chargeback and get back to us in his ticket to receive a refund.
Very good. Just to be clear, a full refund of the entire bw overage charge?
Could you please provide instructions on how to use this feature? I haven’t been able to locate it. Your guidance would be greatly appreciated.